cs.AI（2024-12-05）

📊 共 23 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱一：机器人控制 (Robot Control) (2 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation	深入剖析CoT提示机制：解码、投影与激活视角下的LLM推理能力增强	large language model chain-of-thought
2	TANGO: Training-free Embodied AI Agents for Open-world Tasks	TANGO：无需训练的具身智能体，解决开放世界任务	embodied AI large language model
3	Movie Gen: SWOT Analysis of Meta's Generative AI Foundation Model for Transforming Media Generation, Advertising, and Entertainment Industries	Movie Gen：Meta的生成式AI基础模型在媒体生成、广告和娱乐行业的SWOT分析	foundation model multimodal
4	Leveraging Large Language Models to Generate Course-specific Semantically Annotated Learning Objects	利用大型语言模型生成课程相关的语义标注学习对象	large language model
5	Pre-train, Align, and Disentangle: Empowering Sequential Recommendation with Large Language Models	提出PAD框架，利用预训练语言模型增强序列推荐，解决冷启动和性能瓶颈。	large language model	✅
6	From Code to Play: Benchmarking Program Search for Games Using Large Language Models	利用大型语言模型进行游戏程序搜索的基准测试研究	large language model
7	Automated Multi-Label Annotation for Mental Health Illnesses Using Large Language Models	提出基于大语言模型的多标签自动标注方法以解决心理健康疾病数据不足问题	large language model
8	REL: Working out is all you need	提出REL方法，通过构造高质量推理过程数据提升LLM的规划能力	large language model chain-of-thought
9	Bench-CoE: a Framework for Collaboration of Experts from Benchmark	提出Bench-CoE框架，利用基准评测实现专家模型协同，提升多任务性能	large language model multimodal	✅
10	Exploring AI Text Generation, Retrieval-Augmented Generation, and Detection Technologies: a Comprehensive Overview	全面综述AI文本生成、检索增强生成及检测技术，探讨其发展与伦理影响	large language model
11	MISR: Measuring Instrumental Self-Reasoning in Frontier Models	提出评估工具以测量前沿模型的自我推理能力	large language model	✅
12	Densing Law of LLMs	提出容量密度概念，揭示大语言模型能力随时间指数增长的规律。	large language model
13	PoTable: Towards Systematic Thinking via Stage-oriented Plan-then-Execute Reasoning on Tables	提出PoTable以解决表格推理中的系统性思维问题	large language model
14	Enhancing Mathematical Reasoning in LLMs with Background Operators	利用背景算子增强大型语言模型在数学推理中的能力	large language model
15	Practical Considerations for Agentic LLM Systems	针对Agentic LLM系统，本文从实践角度提出规划、记忆、工具和控制流四大类设计考量。	large language model
16	SocialMind: LLM-based Proactive AR Social Assistive System with Human-like Perception for In-situ Live Interactions	SocialMind：基于LLM的主动AR社交辅助系统，用于实时交互	large language model
17	How Good is ChatGPT in Giving Adaptive Guidance Using Knowledge Graphs in E-Learning Environments?	提出结合知识图谱的自适应指导方法，提升ChatGPT在E-learning环境中的表现	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
18	HEAL: Hierarchical Embedding Alignment Loss for Improved Retrieval and Representation Learning	提出HEAL：通过层级嵌入对齐损失改进检索和表征学习	representation learning contrastive learning large language model
19	Demonstration Selection for In-Context Learning via Reinforcement Learning	提出RDES，利用强化学习进行ICL的demonstration选择，提升泛化性	reinforcement learning PPO chain-of-thought
20	Does your model understand genes? A benchmark of gene properties for biological and text models	提出基因属性基准测试，用于评估生物和文本模型对基因的理解能力	predictive model large language model foundation model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
21	Targeting the Core: A Simple and Effective Method to Attack RAG-based Agents via Direct LLM Manipulation	通过直接操纵LLM核心攻击RAG智能体：一种简单有效的方法	manipulation large language model
22	Monet: Mixture of Monosemantic Experts for Transformers	提出Monet架构以解决大语言模型的多义性问题	manipulation large language model	✅

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
23	EgoPlan-Bench2: A Benchmark for Multimodal Large Language Model Planning in Real-World Scenarios	提出EgoPlan-Bench2基准测试，评估多模态大语言模型在真实场景中的规划能力	egocentric large language model multimodal	✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页